Prioritization of treatment and its consequences: implications of accountability measures [abstract 46-OR] Abstract uri icon
  • At a given office visit, well over half of adults with type 2 diabetes are above goal for glycemia, blood pressure, and lipids. A physician must often decide which clinical domain and action is most likely to lower the risk of complications. Physicians can prioritize treatment decisions at a given patient visit to: treat the condition that is (A) furthest from goal, or (B) closest to goal. Using a simulation model for diabetes management and a simulated population of 1000 patients (modeled on a real population of type 2 diabetes patients), we assessed the effect of these treatment strategies on long-term cardiovascular risk for patients. We used a population percentile metric to determine the distance from goal for each clinical domain, e.g., if a patient is in the 95th percentile for SBP and 80th for LDL, SBP is assumed further from goal. We examined the impact of two diabetes treatment prioritization strategies, one that treats the condition with the highest percentile and the other the lowest percentile over 4 simulated encounters per year for 5 years. Outcomes assessed include (i) % of patients at goal, and (ii) 10-year cardiovascular (CVD) risk estimated using the UKPDS risk engine. As shown in Table 1, rule (B) brings a higher percentage of patients to goal for blood pressure and lipids but not A1c. However, rule (A) achieves a greater reduction in projected CVD events. Results indicate that, if physicians adopt strategies that target the clinical domain that seems easiest to achieve goal, it will maximize percent of patients at goal but may forego an opportunity to lower CHD risk. Diabetes accountability measures that incentivize providers to achieve thresholds for clinical goals, as opposed to lowering cardiovascular risk, may lead to unintended adverse consequences for patients with type 2 diabetes.

    Table 1
    A:Furthest from Goal B:Closest to Goal
    Outcome Initial Yr 1 Yr 3 Yr 5 Yr 1 Yr 3 Yr 5
    Mean A1c (% *) 11.5 9.5(0) 8.0(6) 7.5(58) 10.5(0) 8.9(34) 8.0(57)
    Mean SBP (% *) 171 155(0) 141(16) 136(50) 150(20) 142(60) 138(80)
    Mean LDL (% *) 127 103(42) 93(57) 88(75) 106(72) 100(77) 90(84)
    CVD Events ** 508 368 292 267 394 325 288
    *% at goal; **CVD events are based on 10-year risk.

  • publication date
  • 2011
  • published in
  • Diabetes  Journal
  • Research
  • Cardiovascular Diseases
  • Diabetes
  • Models
  • Risk Factors
  • Simulation
  • Additional Document Info
  • 60
  • issue
  • Suppl 1